Foam Superstabilization by Polymer Microrods

Rossitza G. Alargova,† Devdutta S. Warhadpande,† Vesselin N. Paunov,‡ andOrlin D. Velev*,†

Department of Chemical Engineering, North Carolina State University,Raleigh, North Carolina 27695-7905, and Surfactant & Colloid Group,

Department of Chemistry, University of Hull, Hull, HU6 7RX, United Kingdom

Received June 1, 2004. In Final Form: October 12, 2004

Few foam systems stabilized by solid particles have been reported, and usually the particles have beenused in combination with surfactants. We report how foams can be stabilized solely with a new class ofanisotropic particles, hydrophobic polymer microrods of diameter less than 1 µm and length of a few tensof micrometers. The obtained foams were extraordinary stable, retaining a constant volume over manydays and even surviving drying of most of the free liquid. The bubbles in these foams were stericallystabilized by dense thick “hairy” layers. The rigid intertwined protective shells around the bubbles didnot allow the formation of thin films between them. The lifetime of these foams was orders of magnitudelonger than the ones stabilized with typical foaming surfactants such as sodium dodecyl sulfate. Theaddition of sodium dodecyl sulfate led to hydrophilization of the microrods and suppressed thesuperstabilization effect. Thus, common foaming agents effectively act as defoamers for the ultrastablefoams stabilized by microrods.

Introduction

Foams of different types are used in many products, inindustrial processes, and for synthesis of new materials.1-4

Long-term stability is desirable for many foam-basedsystems, but achieving it is rather difficult. The foamsare both thermodynamically and kinetically unstablesystems that readily destruct as a result of liquid drainage,drying, film breakup and gas diffusion. Typical foamingagents used in industrial formulations are ionic surfac-tants such as sodium dodecyl sulfate (SDS) and surface-active polymers. Foam stabilization may also be promotedor inhibited by the presence of solid particles.1,2,5 In mostof the foams studied previously, the solid particles areused in combination with surfactants. The stability of suchfoams has been found to depend on the particle size, shape,concentration, and hydrophobicity, as well as on the typeof surfactant used.1,2,5 Data and theoretical conclusionsin the literature point out that hydrophobic sphericalparticles may initiate foam destruction by rupturing thefoam films (“lamellae”) via a bridging-dewetting mecha-nism.1,2,5-8 On the contrary, if the particles are hydrophilic,they could collect in the plateau borders slowing down theliquid drainage and kinetically increasing the foamstability. Very few studies of foams stabilized solely bysolid particles are reported,5 which include foams formedby a latex suspension close to its coagulation point5 orshort living foams stabilized by sludge particles.9 A large

number of nano- and microparticles of various sizes andshapes synthesized recently following the thrust innanoscience remain largely unexplored. Foam stabiliza-tion solely by particles may be valuable for applicationsin which surfactants are to be avoided, for synthesis ofadvanced materials, or for formation of ultrastable foams,which is the case described in this Letter.

We report how extremely stable aqueous foams can beformed from suspensions of polymer rodlike particles. Theprocess for synthesis of the new “microrods” used forstabilizing these foams was developed recently in ourgroup.10 We describe the structure of foams and singlefoam films of extraordinary long-term stability. The resultson foamability and stability of the microrod-containingsystems are compared to those of foams stabilized withSDS and its mixture with microrods. We discuss the factorsmaking these rod-stabilized foams of almost infinitelifetime and resistant to drying.

Experimental Section

Polymer Microrod Preparation. The polymer rodlikeparticles used for foam stabilization were synthesized from epoxy-type photoresist SU-8 (MicroChem, MA) using the liquid-liquiddispersion technique described in detail elsewhere.10 Typically,0.2 mL of 50 wt % solution of SU-8 in γ-butyrolactone (GBL)(Aldrich, WI) was injected into 50 mL of a 1/1 v/v mixture ofglycerol and ethylene glycol (Fisher Scientific, PA) subjected toa viscous shearing by an impeller. The polymer solution wasemulsified, and the emulsion droplets deformed and solidifiedinto cylinders with high aspect ratio. The polymer particles werestrengthened by cross-linking with UV light. Then the dispersionwas diluted twice with deionized (DI) water (from a MilliporeRiOs 16 RO unit) and filtered through a 0.45 µm Duraporemembrane (Millipore, MA) to separate the particles. The SU-8rods collected on the membrane were repeatedly washed with DIwater to remove any organic solvent traces and then suspendedinto deionized water to give a dispersion of 2.18 wt % solid content.Microscopy measurements determined that the samples consistedof polydisperse cylindrical rods of 23.5 µm average length and0.60 µm average diameter (Figure 1a and Supporting Informa-tion).

* Corresponding author. E-mail: odvelev@unity.ncsu.edu. Fax:919-515-3465.

† North Carolina State University.‡ University of Hull.(1) Aveyard, R.; Binks, B. P.; Fletcher, P. D. I.; Peck, T. G.; Rutherford,

C. E. Adv. Colloid Interface Sci. 1994, 48, 93.(2) Pugh, R. J. Adv. Colloid Interface Sci. 1996, 64, 67.(3) Foams and Emulsions; Sadoc, J. F., Rivier, N., Eds.; NATO ASI

Series E, Vol. 354; Kluwer Academic: Dordrecht, 1999.(4) Shirtcliffe, N. J.; McHale, G.; Newton, M. I.; Perry, C. C. Langmuir

2003, 19, 5626.(5) Binks, B. P. Curr. Opin. Colloid Interface Sci. 2002, 7, 21.(6) Aveyard, R.; Clint, J. H.; Nees, D. Colloid Polym. Sci. 2000, 278,

155.(7) Ip, S. W.; Wang, Y.; Toguri, J. M. Can. Metall. Q. 1999, 38, 81.(8) Kaptay, G. Colloids Surf., A 2004, 230, 67.(9) Bindal, S. K.; Nikolov, A. D.; Wasan, D. T.; Lambert, D. P.;

Koopman, D. C. Environ. Sci. Technol. 2001, 35, 3941.(10) Alargova, R. G.; Bhatt, K. H.; Paunov, V. N.; Velev, O. D. Adv.

Mater. 2004, 16, 1653.

10371Langmuir 2004, 20, 10371-10374

10.1021/la048647a CCC: $27.50 © 2004 American Chemical SocietyPublished on Web 10/29/2004

Foam Preparation and Characterization. Foams stabi-lized by SU-8 rods only were prepared directly from the abovemicrorod suspension or after its further dilution with DI water.For a reference surfactant system, we used an 8.7 mM solutionof SDS in water. The SDS concentration is close to its criticalmicellar concentration (cmc) where it shows maximum foam-ability.2 All foams were formed by the same protocol from 2 mLof liquid measured by a pipet into a 10 mL glass cylinder. Theywere prepared at room temperature by hand-shaking for a periodof 30 s by the same operator. The foamability was estimated bymeasuring the foam volume, Vf, immediately after preparation,while the foam stability was assessed by monitoring Vf over time.The cylinders were kept closed except for the case when thestability of the rod foam upon drying was evaluated.

The structure of small foam bubbles and thin liquid films wasobserved by an Olympus BX-61 microscope equipped with longworking distance objectives. The bubbles were mounted in athick film on microscope slides. Single foam films were formedinside 1.4 mm holes in a polycarbonate slide of 0.1 mm thicknessby introducing a drop of the SU-8 rod suspension into the holeand sucking out the excess liquid. The observations wereperformed in transmitted phase contrast or dark field illumina-tion.

Results and DiscussionPhenomenology. The microrod suspensions readily

produced foams upon shaking. Batches of foams with fourdifferent concentrations of SU-8 rods were prepared andmonitored. Initially, the air bubbles formed were spreadthroughout the entire volume of the samples, making themmilky. In a few minutes, most of the air bubbles ac-cumulated in the top phase, leaving a less turbid lowerphase of SU-8 rods dispersed in water. The foam volume,Vf, recorded as a function of time for all samples stabilizedby rods only is given in Figure 2a. Vf decreased noticeablyonly within the first few minutes (see the inset) and thenremained approximately the same for more than 3 weeks.The long-term values of Vf as a function of the rodconcentration are presented in Figure 3. They show apossible initial thresholdconcentrationwherestabilization

begins and a rapid increase of the foam volume with theparticle content.

Figure 1. (a) Optical micrograph of the SU-8 microrods usedas superstabilizers; (b) foam stabilized by rods; (c) micrographof a single hairy air bubble covered by a layer of adsorbed rods;(d) single thin aqueous film formed from a suspension of SU-8rods. The inset in image d is a highly magnified area near thefilm center. The scale bar is 50 µm for images a and c and theinset of image d and 200 µm for images b and d.

Figure 2. (a) Time dependence of the volume of foams, Vf,formed with SU-8 rods. The lines of experimental points (lackof any foam breakdown) correspond to different solid concen-trations: ([) 2.18 wt %; (0) 1.09 wt %, (2) 0.44 wt %; (O) 0.22wt %. The inset shows the volume changes for the first 30 min.(b) Time dependence of the volume of foams, Vf, formed in thepresence of (4) 8.7 mM SDS, (2) 2.18 wt % SU-8 rods mixedwith 8.7 mM SDS, and ([) 2.18 wt % SU-8 rods only. The insetshows the volume changes for the first 30 min. The curves areguides to the eye.

Figure 3. Volume of the foam formed as a function of SU-8microrod concentration in the water phase. The data correspondto plateau values for Vf in Figure 2a. The line is a guide to theeye.

10372 Langmuir, Vol. 20, No. 24, 2004 Letters

A more stringent test for foam stability is the resistanceto drying, which typically destroys any surfactant-stabilized foam. The resistance of microrod foam (2.18 wt% SU-8) to water evaporation was examined by monitoringits state in a cylinder open to the air. For more than aweek, the volume of the foam remained at a constant valueof 1.3 mL (similar to the Vf of the same foam kept in aclosed cylinder), even though the liquid below it slowlyevaporated through the foam layer. The evaporationpossibly occurs by capillary suction from the subphase.After a week, we tried to destroy the remaining foam byfast drying and expansion in a vacuum. When the pressurewas reduced, the foam started increasing in volume andreachedtwice its initial volumewhere it remainedconstantfor hours. The major part of the (already dried) foamcollapsed only when the vacuum was released, and airrushed into the vessel. Thus all tests prove an extra-ordinary effect of “superstabilization” by the microrods.

It is instructive to compare the properties and thestructure of microrod-stabilized foams to the ones madefrom common surfactants such as SDS. The stabilities offoams prepared with pure SDS, pure 2.18 wt % SU-8microrods, and a mixture of SDS and 2.18 wt % microrodsare compared in Figure 2b. Despite the large differencein the foamability (the initial Vf is about 5 times largerfor surfactant-stabilized foams), the shape of the short-term stability plot (the inset) is similar for all foams andindicates similar draining changes in the initial period.However, the long-term plots show that the microrod-stabilized foam has strikingly superior stability comparedto the SDS-containing samples, both of which are com-pletely destroyed in less than 48 h. This effect is alsodemonstrated by the images of the fresh and 2-day-oldfoams in Figure 4.

The sample containing mixed SDS and microrods wasexpected to show an intermediate stability; however, itbehaved as pure SDS foam except for the slightly smallerinitial volume and lifetime. This can be readily explainedby the adsorption of SDS monomers on the hydrophobicSU-8 rod surface. These hydrophilized rods lose theiraffinity for adsorption at the solution/air interface and inaddition repel from it electrostatically. Instead of protect-

ing the bubbles, they remain well dispersed in the liquidphase and slowly sediment to the bottom (Figure 4e).Estimates show that if the surface of all rods present ina 2.18 wt % dispersion is covered by densely packed SDSmonomers, the initial surfactant concentration is reducedby only about 5%. These data lead to an unusualconclusion. SDS, which is used typically as a strongfoaming agent, in this case acts as a defoamer suppressingthe stabilizing effect of the microrods. This was also provedby adding gently a few drops of 10 wt % SDS to a stablefew-days-old foam from 1.09 wt % SU-8 rods. Soon afterthe surfactant was introduced, the microrod particlesbegan to transfer from the foam into the liquid and tosediment at the bottom. More than 70% of the foam phasewas destroyed in the first 30 min.

Factors Contributing to the SuperstabilizationEffect. We assume that the initial compression of thefoam volume was a result of liquid drainage and bubblecompaction, rather than loss of any entrapped air. Closerexamination of the microrod foam under a microscopeshowed that it was made of small roughly sphericalrandomly distributed bubbles (Figure 1b). Each air bubblewas covered by a dense shell of adsorbed entangledpolymer rods extending into the water phase (Figure 1c).

The key to making stable foams is ensuring that thethin films formed between the bubbles do not thin andbreak easily.12-14 We examined single foam films in orderto observe the structure of the SU-8 rods adsorbed betweenthe two air/water interfaces. The whole surface of suchfilms was densely covered by intertwined rods (see Figure1d and Supporting Information, Figure 4). The rodsentangle, overlap, and sometimes form small orienteddomains. The film thickness corresponds to at least twoopposing layers of rods, or ca. 1-2 µm. This is anoverwhelmingly thick film compared to the equilibriumthickness of surfactant foam films stabilized by electro-static repulsion (≈100 nm)12,13 or the thickness of common“black” films with steric repulsion between surfactantmonolayers adsorbed on the opposing surfaces (≈12nm).12,13 Thus, the first factor of stability of the super-stabilized foam is the steric repulsion between theadsorbed layers of solid rods, which keeps the films verythick, preventing breakage and suppressing gas diffusion.Even though there are some empty spaces between therods, the capillary pressure in the system would not beenough to locally thin and break the films there.

The second unique factor of superstabilization is themechanical rigidity of the continuous net of overlappingand entangled microrods at the film surface (cf. inset inFigure 1d). The microrods did not rearrange as previouslyobserved with spherical particles entrapped in thin foamfilms.15 The stability of the formed microrod net isimpressive. Similarly to the foam systems, single filmswith concentration above ≈1 wt % were infinitely stableand were not destabilized by drying, which leads to rapidbreaking down of films made with SDS only. The filmsand membranes from intertwined rods could be dried andkept intact for more than a week.

One of the major factors for the formation of thick andrigid films between the bubbles is the strong adsorptionof the polymer rods on the air/water interface. The degree

(11) Paunov, V. N. Langmuir 2003, 19, 7970.(12) Exerowa, D. R.; Krugliakov, P. M. Foam and Foam Films: Theory,

Experiment, Application; Studies in Interface Science, Vol. 5; Elsevier:New York, 1998.

(13) Thin Liquid Films; Ivanov, I. B., Ed.; Surfactant Science Series,Vol. 29; Marcel Dekker: New York, 1988.

(14) Morrison, I. D.; Ross, S. Colloidal Dispersions: Suspensions,Emulsions, Foams; Wiley-Interscience: New York, 2002; Chapter 23.

(15) Velikov, K. P.; Durst, F.; Velev, O. D. Langmuir 1998, 14, 1148.

Figure4. Comparison of the appearance and structure of foamsstabilized by (a,d) 8.7 mM SDS, (b,e) a mixture of 8.7 mM SDSand 2.18 wt % SU-8 rods, and (c,f) 2.18 wt % SU-8 rods. Thetop images (a-c) were obtained 10-15 min after the foam wasformed. The bottom images (d-e) are of the top system after2 days of storage. Note that SDS not only makes much lessstable foams than the microrods but effectively acts as adefoaming agent for the microrod foams.

Letters Langmuir, Vol. 20, No. 24, 2004 10373

of rod hydrophobicity was characterized by measurementsof the contact angle on single rods by a recently developedgel trapping technique.11 The contact angle between SU-8rods and water was estimated to be between 80° and 90°corresponding to intermediate hydrophobicity (see Sup-porting Information). The adsorbed microrods were par-tially immersed in the water phase but did not aggregatespontaneously when dispersed in the bulk. It has beenreported that partially hydrophobic particles are optimalstabilizers of various foam and emulsion films.1,2,15-17 Wealso observed by scanning electron microscopy (SEM)entanglement of the rods adsorbed at the interface(Supporting Information). The strong adhesion betweenthe adjacent and overlapping rods in the films (responsiblefor the rigidity of the adsorbed layer) is probably aug-mented by their hydrophobicity and high friction. The rodentanglement appears to be the major difference betweenthe superstabilized foams reported here and previouslystudied foams containing spherical particles.1,2,15,18 Theeffect is lost when the particles are hydrophylized anddesorbed from the interface due to SDS adsorption, whencethe superb stability is lost when common surfactant isadded.

The bulk structure of the superstabilized foams wasalso very different from that of common foams stabilizedby surfactants. The microrod foams were made of ap-proximately spherical air bubbles (Figure 1b) not only forthe very first stage of the foam formation but for the wholeperiod of observation. Such structure is typical for unstablewet transient foams which contain large amounts of waterand live only seconds.19 These bubbles did not deform toreach the second (kugelschaum) and third (polyeder-schaum) stages of foam evolution2 in which air bubbles incommon foams deform, the liquid from the films betweenthem drains, and a “dry” foam with thin films and plateauborders separating large polyhedral air cells forms (cf.Figure 4a,c). Yet, the rod-stabilized foams show remark-ably long lifetimes exceeding dramatically even thelifetime of the dry metastable foams. The reason for thelack of deformation is probably the rigidity of the dense“hairy” shells around the bubbles. We have recently shownthat rods form similarly rigid shells around emulsiondroplets.20

The effects presented here also differ significantly fromliterature data for other particle-stabilized foams. It hasbeen reported that strongly hydrophobic particles decreasefoam stability by breaking the lamellae by a “bridging-dewetting” mechanism.1,5 Foam stabilization with slightlyhydrophobic and with hydrophilic particles is usually aresult of collecting of the particles in the plateau borders,which slows down the liquid drainage and therefore thethinning and rupture of the lamellae.1,2 In the case ofpolymer microrod foams, the stabilization mechanismincludes the formation of a thick rigid net of entangledrods around the bubbles. The bubbles do not coalescebecause of the strong steric hindrance from their shells.Diffusion of air between the bubbles is not likely to occurbecause the liquid films between them are very thick andsmall18 and because the bubbles covered by a rigid shellcannot readily shrink or expand. The strong rod adsorptionand entanglement, formation of rigid hairy shells, andsustaining of thick films makes possible the superstabi-lization effect.

Concluding Remarks. We have shown that super-stabilization of aqueous foams can be achieved by usingpolymer microrods in the absence of any surfactant. Dueto their rigid structure and resistance to mechanicalperturbations, the polymer rodlike foams can be used inapplications where the common foam stabilizers are noteffective. Foams based on polymer microrods and theirmixtures or composites with other nano- and micropar-ticles may also serve for synthesis of new materials. In amore general plan, we believe that this work shows howprogress in the area of controlled synthesis of newnanoparticles and nanostructures can dramatically en-hance the control over the properties of “classical” colloidalsystems such as foams and emulsions.

Acknowledgment. This study was supported by NERand CAREER grants from the National Science Founda-tion (U.S.) and by EPSRC Grant GR/S10025/01 (U.K.).

Supporting Information Available: SEM and opticalmicrographs and size distribution of the microrod samples; SEMimages of polymer cylinders adsorbed at replicas of air-waterinterfaces. This material is available free of charge via theInternet at http://pubs.acs.org.

LA048647A

(16) Kruglyakov, P. M.; Nushtayeva, A.; Vilkova, N. G. J. Colloid.Interface Sci. 2004, 276, 465.

(17) Kruglyakov, P. M.; Nushtayeva, A. V. Adv. Colloid InterfaceSci. 2004, 108, 151.

(18) Du, Z. P.; Bilbao-Montoya, M. P.; Binks, B. P.; Dickinson, E.;Ettelaie, R.; Murray, B. S. Langmuir 2003, 19, 3106.

(19) Malysa, K. Adv. Colloid Interface Sci. 1992, 40, 37.(20) Noble, P. F.; Cayre, O. J.; Alargova, R. G.; Velev, O. D.; Paunov,

V. N. J. Am. Chem. Soc. 2004, 126, 8092.

10374 Langmuir, Vol. 20, No. 24, 2004 Letters